Anisotropic shearing mechanism of Kangding slate:Experimental investigation and numerical analysis

The shear mechanical behavior is regarded as an essential factor affecting the stability of the surrounding rocks in underground engineering.The shear strength and failure mechanisms of layered rock are significantly affected by the foliation angles.Direct shear tests were conducted on cubic slate samples with foliation angles of 0°,30°,45°,60°,and 90°.The effect of foliation angles on failure patterns,acoustic emission(AE)characteristics,and shear strength parameters was analyzed.Based on AE characteristics,the slate failure process could be divided into four stages:quiet period,step-like increasing period,dramatic increasing period,and remission period.A new empirical expression of cohesion for layered rock was proposed,which was compared with linear and sinusoidal cohesion expressions based on the results made by this paper and previous experiments.The comparative analysis demonstrated that the new expression has better prediction ability than other expressions.The proposed empirical equation was used for direct shear simulations with the combined finite-discrete element method(FDEM),and it was found to align well with the experimental results.Considering both computational efficiency and accuracy,it was recommended to use a shear rate of 0.01 m/s for FDEM to carry out direct shear simulations.To balance the relationship between the number of elements and the simulation results in the direct shear simulations,the recommended element size is 1 mm.

Research on Active and Reactive Power Decoupling Control Based on VSC-HVDC

Voltage Source Converter-based High Voltage Direct Current(VSC-HVDC)transmission technology represents a groundbreaking approach in high voltage Direct Current(DC)transmission,offering numerous technical advantages and broad application prospects.However,in the d-q synchronous rotating coordinate system,the VSC-HVDC exhibits the coupling effect of active power and reactive power,so it needs to be decoupled.This paper introduces the basic principle and mathematical model of the VSC-HVDC transmission system.Through the combination of coordinate transformation and variable substitution,a feedforward decoupling control method is derived.Then the VSC-HVDC simulation model is designed,and the simulation analysis is carried out in the MATLAB environment.The simulation results demonstrate that the method effectively achieves decoupling control of active and reactive power,exhibiting superior dynamic performance and robustness.These findings validate the correctness and effectiveness of the control strategy.

Robust state of charge estimation of lithium-ion battery via mixture kernel mean p-power error loss LSTM with heap-based-optimizer

The state of charge(SOC)estimation of lithium-ion battery is an important function in the battery management system(BMS)of electric vehicles.The long short term memory(LSTM)model can be employed for SOC estimation,which is capable of estimating the future changing states of a nonlinear system.Since the BMS usually works under complicated operating conditions,i.e the real measurement data used for model training may be corrupted by non-Gaussian noise,and thus the performance of the original LSTM with the mean square error(MSE)loss may deteriorate.Therefore,a novel LSTM with mixture kernel mean p-power error(MKMPE)loss,called MKMPE-LSTM,is developed by using the MKMPE loss to replace the MSE as the learning criterion in LSTM framework,which can achieve robust SOC estimation under the measurement data contaminated with non-Gaussian noises(or outliers)because of the MKMPE containing the p-order moments of the error distribution.In addition,a meta-heuristic algorithm,called heap-based-optimizer(HBO),is employed to optimize the hyper-parameters(mainly including learning rate,number of hidden layer neuron and value of p in MKMPE)of the proposed MKMPE-LSTM model to further improve its flexibility and generalization performance,and a novel hybrid model(HBO-MKMPE-LSTM)is established for SOC estimation under non-Gaussian noise cases.Finally,several tests are performed under various cases through a benchmark to evaluate the performance of the proposed HBO-MKMPE-LSTM model,and the results demonstrate that the proposed hybrid method can provide a good robustness and accuracy under different non-Gaussian measurement noises,and the SOC estimation results in terms of mean square error(MSE),root MSE(RMSE),mean absolute relative error(MARE),and determination coefficient R2are less than 0.05%,3%,3%,and above 99.8%at 25℃,respectively.

内建电场辅助光催化甲烷干重整的研究进展

甲烷(CH_(4))和二氧化碳(CO_(2))是导致全球变暖的两种主要温室气体.甲烷干重整技术能够同时消耗两种温室气体并制备氢气(H2)和一氧化碳(CO),是减少温室效应的理想策略之一.CH_(4)和CO_(2)在热力学上具有很高的稳定性,所以活化CH_(4)和CO_(2)需要克服较高的能垒,导致传统的甲烷干重整技术总是需要高热能来触发该反应发生.光催化技术的发展为在温和条件下启动甲烷干重整反应提供了更多的可能.然而,由于光激发载流子之间的快速重组,光催化效率仍然较低,难以满足工业需求.研究人员发现,通过构建内置电场增强电荷载流子的分离和转移动力学是解决上述问题的可靠策略.本文首先介绍了甲烷干重整的反应机理和用于甲烷干重整的工业热催化材料.随后,总结了光催化甲烷干重整的优点和潜在的光催化材料,重点介绍了两类催化剂:(1)由铁电效应产生的永久自发极化进而构筑的内建电场的光催化剂.由于自发极化引起的电场,基于铁电材料的光催化剂在促进电荷转移方面显示出较大潜力.(2)由异质结结构在界面处引发内建电场的光催化剂.基于两种具有合适能带结构的半导体构建Ⅱ型异质结也是一种有效方法,由于交错间隙结构,在界面处形成内置电场,导致不同半导体分别进行氧化和还原过程.此外,Z型载流子转移机制可以保留具有更强还原能力的电子和更强氧化能力的空穴,将较低氧化还原能力的光生载流子重组,从而通过界面电场促进光催化甲烷干重整过程.(3)局域表面等离激元共振(LSPR)效应引发内建热电场的光催化剂.金属纳米颗粒在可见-近红外(Vis-NIR)光的照射下会产生共振现象,将会导致金属中的电子结构不连续,从而构建局部电场.因此,LSPR效应在提高光(热)催化甲烷干重整效率方面具有巨大潜力.随着光催化甲烷干重整技术的发展,人们对理解反应机理或阐明光催化剂中特定组分在反应中的作用提出了更多要求,导致原位表征技术和理论计算受到了极大的关注.最后,介绍了先进的原位表征和理论计算在该领域应用的主要进展,并预测了原位表征在光催化甲烷干重整领域的潜在功能,为从事该领域且处于起步阶段的年轻研究者提供了一定参考.虽然在光催化甲烷干重整领域已经取得了许多突破和进展,但仍存在一些挑战需要克服.根据已有的研究结果,本文总结了内建电场辅助光催化甲烷干重整领域的主要面临挑战,并提出了应对这些挑战的可行性策略,为未来对该领域进行更深入的研究提供借鉴.

Focal adhesion protein Kindlin-2 regulates bone homeostasis in mice

Our recent studies demonstrate that the focal adhesion protein Kindlin-2 is critical for chondrogenesis and early skeletal development. Here, we show that deleting Kindlin-2 from osteoblasts using the 2.3-kb mouse Col1 a1-Cre transgene minimally impacts bone mass in mice, but deleting Kindlin-2 using the 10-kb mouse Dmp1-Cre transgene, which targets osteocytes and mature osteoblasts, results in striking osteopenia in mice. Kindlin-2 loss reduces the osteoblastic population but increases the osteoclastic and adipocytic populations in the bone microenvironment. Kindlin-2 loss upregulates sclerostin in osteocytes,downregulates β-catenin in osteoblasts, and inhibits osteoblast formation and differentiation in vitro and in vivo. Upregulation ofβ-catenin in the mutant cells reverses the osteopenia induced by Kindlin-2 deficiency. Kindlin-2 loss additionally increases the expression of RANKL in osteocytes and increases osteoclast formation and bone resorption. Kindlin-2 deletion in osteocytes promotes osteoclast formation in osteocyte/bone marrow monocyte cocultures, which is significantly blocked by an anti-RANKLneutralizing antibody. Finally, Kindlin-2 loss increases osteocyte apoptosis and impairs osteocyte spreading and dendrite formation.Thus, we demonstrate an important role of Kindlin-2 in the regulation of bone homeostasis and provide a potential target for the treatment of metabolic bone diseases.

LIM domain proteins Pinch1/2 regulate chondrogenesis and bone mass in mice

The LIM domain-containing proteins Pinch1/2 regulate integrin activation and cell–extracellular matrix interaction and adhesion.Here,we report that deleting Pinch1 in limb mesenchymal stem cells(MSCs)and Pinch2 globally(double knockout;dKO)in mice causes severe chondrodysplasia,while single mutant mice do not display marked defects.Pinch deletion decreases chondrocyte proliferation,accelerates cell differentiation and disrupts column formation.Pinch loss drastically reduces Smad2/3 protein expression in proliferative zone(PZ)chondrocytes and increases Runx2 and Col10a1 expression in both PZ and hypertrophic zone(HZ)chondrocytes.Pinch loss increases sclerostin and Rankl expression in HZ chondrocytes,reduces bone formation,and increases bone resorption,leading to low bone mass.In vitro studies revealed that Pinch1 and Smad2/3 colocalize in the nuclei of chondrocytes.Through its C-terminal region,Pinch1 interacts with Smad2/3 proteins.Pinch loss increases Smad2/3 ubiquitination and degradation in primary bone marrow stromal cells(BMSCs).Pinch loss reduces TGF-β-induced Smad2/3 phosphorylation and nuclear localization in primary BMSCs.Interestingly,compared to those from single mutant mice,BMSCs from dKO mice express dramatically lower protein levels ofβ-catenin and Yap1/Taz and display reduced osteogenic but increased adipogenic differentiation capacity.Finally,ablating Pinch1 in chondrocytes and Pinch2 globally causes severe osteopenia with subtle limb shortening.Collectively,our findings demonstrate critical roles for Pinch1/2 and a functional redundancy of both factors in the control of chondrogenesis and bone mass through distinct mechanisms.

Meta Ordinal Regression Forest for Medical Image Classification With Ordinal Labels

The performance of medical image classification has been enhanced by deep convolutional neural networks(CNNs),which are typically trained with cross-entropy(CE)loss.However,when the label presents an intrinsic ordinal property in nature,e.g.,the development from benign to malignant tumor,CE loss cannot take into account such ordinal information to allow for better generalization.To improve model generalization with ordinal information,we propose a novel meta ordinal regression forest(MORF)method for medical image classification with ordinal labels,which learns the ordinal relationship through the combination of convolutional neural network and differential forest in a meta-learning framework.The merits of the proposed MORF come from the following two components:A tree-wise weighting net(TWW-Net)and a grouped feature selection(GFS)module.First,the TWW-Net assigns each tree in the forest with a specific weight that is mapped from the classification loss of the corresponding tree.Hence,all the trees possess varying weights,which is helpful for alleviating the tree-wise prediction variance.Second,the GFS module enables a dynamic forest rather than a fixed one that was previously used,allowing for random feature perturbation.During training,we alternatively optimize the parameters of the CNN backbone and TWW-Net in the meta-learning framework through calculating the Hessian matrix.Experimental results on two medical image classification datasets with ordinal labels,i.e.,LIDC-IDRI and Breast Ultrasound datasets,demonstrate the superior performances of our MORF method over existing state-of-the-art methods.

A Networking Solution on Uplink Channel of CoFrequency and Co-Time System

Co-frequency and co-time full duplex(CCFD) is an attractive technology for the future wireless communication because of its high spectral efficiency.However,applications of CCFD to mobile network can suffer from strong base station to base station(B2B)interference.In this paper,the authors proposed a design that uses centralized base station(BS)transmit antenna and distributed BS receive antennas,each of which consists of an antennary to perform beamforming that can nullify the B2 B interference.In addition,we proposed a combination algorithm that uses the zero forcing method to cascade the recursive least square(RLS) method for reducing the necessary number of the bits taken to the digital processor.This enables the faster convergence and,thus,allows the transmission of more information bits,compared to the conventional method,for mobile communication.The simulation results confirm this approach for practical application.

Future of Artificial Intelligence in Anesthetics and Pain Management

The potential of the second wave of Artificial Intelligence (AI) to change our lives beyond recognition is both exciting and challenging. AI has been around for over three decades, and this new approach of artificial intelligence, due to enhancements in technology, both software, and hardware, has resulted in the fact that human decision-making is considered inferior and erratic in many fields: none more so than medicine. Machine learning algorithms with access to large data sets can be trained to outperform clinicians in many respects. AI’s effectiveness in accurate diagnosis of various medical conditions and medical image interpretation is well documented. Modern AI technology has the potential to transform medicine to a level never seen before in terms of efficiency and accuracy;but is also potentially highly disruptive, creating insecurity and allowing the transfer of expert domain knowledge to machines. Anesthetics is a complex medical discipline and assuming AI can easily replace experienced and knowledgeable medical practitioners is a very unrealistic expectation. AI can be used in anesthetics to develop, in some respects, more advanced clinical decision support tools based on machine learning. This paper focuses on the complexity of both AI developments, deep learning, neural networks, etc. and opportunities of AI in anesthetics for the future. It will review current advances in AI tools and hardware technologies as well as outlining how these can be used in the field of anesthetics.

Prompt learning in computer vision: a survey

Prompt learning has attracted broad attention in computer vision since the large pre-trained visionlanguagemodels (VLMs) exploded. Based on the close relationship between vision and language information builtby VLM, prompt learning becomes a crucial technique in many important applications such as artificial intelligencegenerated content (AIGC). In this survey, we provide a progressive and comprehensive review of visual promptlearning as related to AIGC. We begin by introducing VLM, the foundation of visual prompt learning. Then, wereview the vision prompt learning methods and prompt-guided generative models, and discuss how to improve theefficiency of adapting AIGC models to specific downstream tasks. Finally, we provide some promising researchdirections concerning prompt learning.

Accelerating the design of multicomponent rare earth silicates for SiC_(f)/SiC CMC by combinatorial material chip design and high throughput screening

High throughput experimentation is employed to establish a ternary system with the compositional range of 30.8 mol.%-75.7 mol.%SiO_(2),16.6 mol.%-61.7 mol.%Yb_(2)O_(3),and 6.3 mol.%-4.1 moll.%Ho_(2)O_(3) through co-sputtering deposition on one combinatorial material chip.Considering their application in advanced SiC_(f)/SiC CMC,the phase composition and mechanical properties of samples with various RE/Si ratios and Yb/Ho ratios are comprehensively investigated.Chemical stability and thermal expansion compatibility between SiC and RE silicates with different compositions are also validated.Optimized materials for the application of environmental barrier coating and interphase for SiC_(f)/SiC CMC are screened respectively according to the above trends and data.This work is a case study to establish a composition-property library for RE_(2)O_(3)-SiO_(2) compounds.It is inspired more complicated multicomponent RE silicates could be prepared and characterized by high throughput experimentation,accelerating the design and screening of promising optimal candidates.

High porosity and low thermal conductivity high entropy(Zr0.2Hf0.2Ti0.2Nb0.2Ta0.2C

Porous ultra-high temperature ceramics(UHTCs)are promising for ultrahigh-temperature thermal insulation applications.However,the main limitations for their applications are the high thermal conductivity and densification of porous structure at high temperatures.In order to overcome these obstacles,herein,porous high entropy(Zr(0.2)Hf(0.2)Ti(0.2)Nb(0.2)Ta(0.2))C was prepared by a simple method combing in-situ reaction and partial sintering.Porous high entropy(Zr(0.2)Hf(0.2)Ti(0.2)Nb(0.2)Ta(0.2))C possesses homogeneous microstructure with grain size in the range of 100–500 nm and pore size in the range of 0.2–1μm,which exhibits high porosity of 80.99%,high compressive strength of 3.45 MPa,low room temperature thermal conductivity of 0.39 W·m^-1K^-1,low thermal diffusivity of 0.74 mm^2·s^-1and good high temperature stability.The combination of these properties renders porous high entropy(Zr(0.2)Hf(0.2)Ti(0.2)Nb(0.2)Ta(0.2))Cpromising as light-weight ultrahigh temperature thermal insulation materials.

Mechanical and thermal properties of RE4Hf3O12（RE=Ho,Er, Tm） ceramics with defect fluorite structure

The thermal and environmental barrier coatings (T/EBC) are technologically important for advanced propulsion engine system. In this study, RE4Hf3Oi2 (RE=Ho, Er, Tm) with defect fluorite structure was investigated for potential use as top TBC layer. Dense pellets were fabricated via a hot pressing method and the mechanical and thermal properties were characterized. RE4Hf3Oi2 (RE=Ho, Er, Tm) possessed a high Vickers hardness of 11 GFa. The material retained high elastic modulus at elevated temperatures up to 1773 K, which made it attractive for high temperature application. The coefficient of thermal expansion (CTE) of RE4Hf3Oi2 (RE = Ho, Er, Tm) laid in the range between 7× 10^-6K^-1 to 10×10^16K^-1 from 473 K to 1673 K. In addition, the rare earth hafnates exhibited lower thermal conductivity which rendered it a good candidate material for thermal barrier applications.

Strided Self-Supervised Low-Dose CT Denoising for Lung Nodule Classification

Lung nodule classification based on low-dose computed tomography(LDCT)images has attracted major attention thanks to the reduced radiation dose and its potential for early diagnosis of lung cancer from LDCT-based lung cancer screening.However,LDCT images suffer from severe noise,largely influencing the performance of lung nodule classification.Current methods combining denoising and classification tasks typically require the corresponding normal-dose CT(NDCT)images as the supervision for the denoising task,which is impractical in the context of clinical diagnosis using LDCT.To jointly train these two tasks in a unified framework without the NDCT images,this paper introduces a novel self-supervised method,termed strided Noise2Neighbors or SN2N,for blind medical image denoising and lung nodule classification,where the supervision is generated from noisy input images.More specifically,the proposed SN2N can construct the supervision infor-mation from its neighbors for LDCT denoising,which does not need NDCT images anymore.The proposed SN2N method enables joint training of LDCT denoising and lung nodule classification tasks by using self-supervised loss for denoising and cross-entropy loss for classification.Extensively experimental results on the Mayo LDCT dataset demonstrate that our SN2N achieves competitive performance compared with the supervised learning methods that have paired NDCT images as supervision.Moreover,our results on the LIDC-IDRI dataset show that the joint training of LDCT denoising and lung nodule classification significantly improves the performance of LDCT-based lung nodule classification.